We have recently designed and synthesized a TR ligand (GC-1) representative of a new class of thyromimetics that has several synthetic advantages over traditional thyromimetics and interesting, potentially useful, thyromimetic properties. Currently, no high affinity antagonist is known for TR, but such a ligand could prove very useful in medicine for treatment of hyperthyroidism and, possibly, cardiac arrhytmia, as well as the study of thyroid hormone signaling pathways. A comparison of the structures of many nuclear receptor agonists and antagonists shows an interesting trend: the antagonists all have the general skeleton of the corresponding agonists plus a large "extension" protruding from the approximate center of the molecule. The recently solved crystal structures of the estrogen receptor complexed with the ER agonist estradiol and with the antagonist raloxifene show that, as a result of the presence of the extension, raloxifene binding impairs receptor folding and prevents the agonist-induced conformational changes in the receptor. The GC-1 thyromimetic scaffold is well suited to generate analogues, several sites are especially amenable of derivatization chemistry. The crystal structures of the ligand binding domains of both TR subtypes (alpha and beta) have been solved. Visual inspection of these structures with the program MidasPlus and using the Computer Graphics Laboratory facilities play an essential role in our effort to make a TR antagonist.